CN111310671B - Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning - Google Patents
Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning Download PDFInfo
- Publication number
- CN111310671B CN111310671B CN202010101729.3A CN202010101729A CN111310671B CN 111310671 B CN111310671 B CN 111310671B CN 202010101729 A CN202010101729 A CN 202010101729A CN 111310671 B CN111310671 B CN 111310671B
- Authority
- CN
- China
- Prior art keywords
- water accumulation
- deep learning
- accumulation pit
- pit
- heating furnace
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Abstract
The invention discloses a heating furnace bottom water accumulation pit anomaly identification method based on deep learning, which comprises the steps of obtaining a water accumulation pit image; detecting the water accumulation pit image by using a trained target detection model to obtain a detection result; the detection result comprises a signboard and a non-signboard; judging whether the water accumulation pit is abnormal or not according to the detection result, if the detection result is that the water accumulation pit comprises a signboard, judging that the water accumulation pit is not abnormal, otherwise, judging that the water accumulation pit is abnormal. According to the invention, manual participation is not needed, the work efficiency is greatly improved for the prior heating furnace bottom water accumulation pit abnormality identification method mainly relying on manual visual inspection, the problem of low efficiency of the existing heating furnace bottom water accumulation pit abnormality identification is solved through machine learning of the characteristics of the identification plate, and a series of problems caused by manual participation are avoided.
Description
Technical Field
The invention relates to the field of digital image processing, in particular to a heating furnace bottom water accumulation pit abnormality identification method, system and equipment based on deep learning.
Background
In the hot rolling process in the ferrous metallurgy field, a heating furnace is required to heat the steel. A water accumulation pit is arranged at the bottom of the heating furnace and is used for collecting water in the production process, and once the water in the water accumulation pit overflows, the water must be treated in time.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the present invention aims to provide a method, a system and a device for identifying abnormal water accumulation pit at the bottom of a heating furnace based on deep learning, which are used for solving the drawbacks of the prior art.
In order to achieve the above and other related objects, the present invention provides a method for identifying abnormal water accumulation pit at bottom of heating furnace based on deep learning, comprising:
acquiring a water accumulation pit image;
detecting the water accumulation pit image by using a trained target detection model to obtain a detection result; the detection result comprises the identification plate and the identification plate which is not included.
Judging whether the water accumulation pit is abnormal or not according to the detection result, if the detection result is that the water accumulation pit comprises a signboard, judging that the water accumulation pit is not abnormal, otherwise, judging that the water accumulation pit is abnormal.
Optionally, the method for training the target detection model comprises:
acquiring a training sample set;
training the neural network based on deep learning based on the training sample set to obtain the target detection model.
Optionally, preprocessing the data in the training sample set, the preprocessing including at least one of: normalization processing and data enhancement;
the data enhancement includes at least one of: clipping, flipping, rotation, change in brightness, change in contrast, change in saturation.
Optionally, the neural network based on deep learning is trained by using a supervised training method.
Optionally, the data in the training sample set is subjected to feature extraction by using the neural network based on deep learning to obtain a feature map, and then the feature map is detected to obtain a detection result.
Optionally, in the process of extracting the feature map, the neural network based on deep learning performs inversion residual error processing and linear bottleneck processing on the data in the training sample set.
Optionally, the neural network based on deep learning is a MobileNetV2-SSD.
Optionally, in training the target detection model, the relu_6 function is used as an activation function, using L2 regularization.
To achieve the above and other related objects, the present invention provides a heating furnace bottom water sump anomaly identification system based on deep learning, comprising:
the image acquisition module is used for acquiring a water accumulation pit image;
the result detection module is used for detecting the water accumulation pit image by using a trained target detection model based on deep learning to obtain a detection result;
and the abnormality judging module is used for judging whether the water accumulation pit is abnormal or not according to the detection result.
To achieve the above and other related objects, the present invention provides an apparatus comprising: a processor and a memory;
the memory is used for storing a computer program, and the processor is used for executing the computer program stored in the memory so as to enable the device to execute the method.
As described above, the heating furnace bottom water accumulation pit abnormality identification method, system and equipment based on deep learning have the following beneficial effects:
according to the invention, manual participation is not needed, the work efficiency is greatly improved for the prior heating furnace bottom water accumulation pit abnormality identification method mainly relying on manual visual inspection, the problem of low efficiency of the existing heating furnace bottom water accumulation pit abnormality identification is solved through machine learning of the characteristics of the identification plate, and a series of problems caused by manual participation are avoided.
Drawings
FIG. 1 is a flow chart of a method for identifying abnormal water accumulation pit at bottom of heating furnace based on deep learning according to an embodiment of the invention;
FIG. 2 is a flow chart of a method for training a target detection model according to an embodiment of the invention;
FIG. 3 is a diagram illustrating a network structure of MobileNet V2 according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a specific network structure of a MobileNetV2 according to another embodiment of the present invention;
FIG. 5 is a schematic diagram of a specific network structure of an SSD according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a heating furnace bottom pit anomaly identification system based on deep learning according to an embodiment of the invention.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complicated.
As shown in fig. 1, the embodiment provides a method for identifying abnormal water accumulation pit at bottom of heating furnace based on deep learning, comprising:
s11, acquiring a water accumulation pit image;
s12, detecting the water accumulation pit image by using a trained target detection model to obtain a detection result; the detection result comprises a signboard and a non-signboard;
s13, judging whether the water accumulation pit is abnormal according to the detection result, if the detection result is that the water accumulation pit comprises the identification plate, judging that the water accumulation pit is not abnormal, otherwise, judging that the water accumulation pit is abnormal.
The method for intelligently detecting the state of the water accumulation pit at the bottom of the heating furnace in real time through the image algorithm can timely detect the overflow of water in the water accumulation pit and timely alarm to remind operators to process.
In one embodiment, as shown in FIG. 2, a method of training a target detection model includes:
s21, acquiring a training sample set;
and S22, training the neural network based on the deep learning based on the training sample set to obtain the target detection model.
And obtaining a clear image of the signboard in the scene, wherein the signboard is moderate in size, and the color can be selected from red or green and the like, so that the signboard has the characteristics of bright color, striking light reflection and good effect. And labeling the image data with the condition that the image data contains the identification plate as normal and the condition that the image data does not contain the identification plate as abnormal, so as to obtain an image data set. Dividing the image data set according to a proportion to obtain a test data set and a training sample set; the proportion of division may be selected from row settings, for example, 1: and 9, taking one tenth of the image data set as a test data set and nine tenth as a training sample set.
After the target detection model is trained, the target detection model is tested by using the test data set. And generating a plurality of candidate detection models in the test process, and taking the candidate detection model with the optimal test result as a target detection model.
In an embodiment, the data in the training sample set is preprocessed, the preprocessing including at least one of: normalization processing and data enhancement;
the normalization processing specifically refers to normalizing the gradation value of the picture from 0 to 255 to 0 to 1. The image normalization uses a maximum and minimum normalization method, and the formula is as follows:
where xi represents the image pixel point value max (x), and min (x) represents the maximum and minimum values of the image pixels, respectively.
The data enhancement includes at least one of: clipping, flipping, rotation, change in brightness, change in contrast, change in saturation.
In one embodiment, a supervised training method is employed to train the deep learning based neural network.
In an embodiment, the data in the training sample set is extracted by using the neural network based on deep learning to obtain a feature map, and then the feature map is detected to obtain a detection result.
In the process of extracting the feature map, the neural network based on deep learning carries out inversion residual error processing and linear bottleneck processing on the data in the training sample set.
In an embodiment, the deep learning based neural network is a MobileNetV2-SSD.
Mobilenet v2 is an improvement over mobilenet v1, which proposes two new concepts: inverted residual Inverted Residual and linear bottleneck Linear Bottleneck. The inverted residual Inverted Residual is mainly used to increase the extraction of image features to improve accuracy, while the linear bottleneck Linear Bottleneck is mainly used to avoid the information loss of the nonlinear function ReLU. The core of mobilenet v2 consists of 17 boltlenecks whose network structure is shown in table 1 (where t is the multiple of the up-scaling in the Bottleneck layer, c is the dimension of the output feature, n is the number of repetitions, s is the step size of the convolution, and k is the width scaling factor).
Table 1 MobileNetV2 network structure table
The specific structure of the bottleneck layer is shown in table 2. The input increases the dimension from k dimension to tk dimension through the 1 x 1 conv+relu layer, then downsamples the image by 3 x 3conv+relu separable convolution (when stride > 1), where the feature dimension is already tk dimension, and finally decreases the dimension from tk to k' dimension by 1 x 1conv (no ReLU).
Table 2 detailed structure table of bottleneck layer
Furthermore, for the bottleneck layer, when stride=1, the sum of elementwise is used to connect the input and output features (as in fig. 3); when stride=2, no shortcut connects the input and output features (as in fig. 4).
SSD is a single-stage object detection algorithm that utilizes feature maps of different scales to predict objects of different frame sizes. The SSD network structure is divided into two parts: basic network + pyramid network, wherein the basic network is transformable. The base network of the original SSD is the first 4-layer network of VGG-16, and the pyramid network is a simple convolution network with a feature map that tapers down, consisting of 5 parts. The specific network structure of the SSD is shown in FIG. 5.
The MobileNet V2-SSD deep learning neural network uses the MobileNet V2 network to replace VGG-16 in the original SSD network architecture, the configuration from Conv0 to Conv13 is completely consistent with the MobileNet V2 model, except that the final global average pooling, full connection layer and Softmax layer of the MobileNet V2 are removed, and the FC6 and FC7 of the original VGG-16 are replaced by Conv6 and Conv7 respectively. The MobileNet V2-SSD deep learning neural network firstly uses the MobileNet V2 network to extract image characteristic output characteristic graphs, and then uses an SSD target detection algorithm to detect information on a plurality of characteristic graphs output by the MobileNet V2 network.
In one embodiment, the ReLU_6 function is used as an activation function in training the target detection model.
To avoid overfitting, the network learning rate is set by adopting an exponential decay method, and an L2 regularization method is adopted (L2 regularization refers to square sums of elements in weight vectors and then square roots of the elements) based on an L2 norm, namely, adding an L2 norm sum term of parameters, namely, square sums of parameters and product terms of parameters, namely, the product terms of the square sums of parameters, to an objective function:
wherein C is 0 Representing the original cost function, n is the number of samples, λ is the regularization term coefficient, and the regularization term is weighted against C 0 The specific gravity of the term, w, is the weight. The term following the plus sign in the formula is the L2 regular term.
In the L2 regularization, the model parameters are updated using the following formula:
in this embodiment, in the target detection process, the image is supervised and trained, and each signboard image has a corresponding tag and a prediction frame, so that model parameters are trained according to the tag and the prediction frame, and the final recognition accuracy is judged according to the tag and the prediction frame. When the training network passes through a plurality of iterations, the predicted value continuously converges to the error direction of the label and the predicted frame, and then the back propagation updates the parameters into each layer according to the chain rule. And each iteration can reduce propagation errors as much as possible according to the gradient descent optimization direction, and finally, the final target detection result of all the signboard images in the data set is obtained. In the invention, a model with highest target detection accuracy on a test set is taken as a target detection model, in the actual operation of an industrial scene, firstly, a picture is acquired in real time through a camera, a single-area water pit signboard image is taken as an input, the model automatically processes the image to identify the characteristics of the signboard, the characteristics of the signboard are predicted, and finally, a target detection result is output. According to the condition of the target detection model identification signboard, whether the water accumulation pit at the bottom of the heating furnace is abnormal or not is judged. And if the final model identifies the water accumulation pit signboard, the explanation is normal. If the final model does not recognize the water accumulation pit signboard, the water accumulation pit signboard is abnormal, and the water overflow in the water accumulation pit at the bottom of the heating furnace is indicated, so that the timely alarm of the abnormal condition of the water accumulation pit at the bottom of the heating furnace is realized.
The method for identifying the abnormal water accumulation pit at the bottom of the heating furnace based on deep learning realizes the identification of the abnormal water accumulation pit at the bottom of the heating furnace in an industrial scene without human participation, has the identification accuracy of more than 99 percent, has excellent effect in the industrial scene of actual heating furnace steelmaking, and has unprecedented leaps in the technical field of identifying the abnormal water accumulation pit at the bottom of the heating furnace.
As shown in fig. 6, a heating furnace bottom water accumulation pit abnormality recognition system based on deep learning includes:
an image acquisition module 11 for acquiring a water accumulation pit image;
the result detection module 12 is configured to detect the sump image by using a trained target detection model based on deep learning, so as to obtain a detection result;
and the abnormality judging module 13 is used for judging whether the water accumulation pit is abnormal according to the detection result.
The method for intelligently detecting the state of the water accumulation pit at the bottom of the heating furnace in real time through the image algorithm can timely detect the overflow of water in the water accumulation pit and timely alarm to remind operators to process.
In one embodiment, a method of training a target detection model includes:
acquiring a training sample set;
training the neural network based on deep learning based on the training sample set to obtain the target detection model.
And obtaining a clear image of the signboard in the scene, wherein the signboard is moderate in size, and the color can be selected from red or green and the like, so that the signboard has the characteristics of bright color, striking light reflection and good effect. And labeling the image data with the condition that the image data contains the identification plate as normal and the condition that the image data does not contain the identification plate as abnormal, so as to obtain an image data set. Dividing the image data set according to a proportion to obtain a test data set and a training sample set; the proportion of division may be selected from row settings, for example, 1: and 9, taking one tenth of the image data set as a test data set and nine tenth as a training sample set.
After the target detection model is trained, the target detection model is tested by using the test data set. And generating a plurality of candidate detection models in the test process, and taking the candidate detection model with the optimal test result as a target detection model.
In an embodiment, the data in the training sample set is preprocessed, the preprocessing including at least one of: normalization processing and data enhancement;
the normalization processing specifically refers to normalizing the gradation value of the picture from 0 to 255 to 0 to 1. The image normalization uses a maximum and minimum normalization method, and the formula is as follows:
where xi represents the image pixel point value max (x), and min (x) represents the maximum and minimum values of the image pixels, respectively.
The data enhancement includes at least one of: clipping, flipping, rotation, change in brightness, change in contrast, change in saturation.
In one embodiment, a supervised training method is employed to train the deep learning based neural network.
In an embodiment, the data in the training sample set is extracted by using the neural network based on deep learning to obtain a feature map, and then the feature map is detected to obtain a detection result.
In the process of extracting the feature map, the neural network based on deep learning carries out inversion residual error processing and linear bottleneck processing on the data in the training sample set.
In an embodiment, the deep learning based neural network is a MobileNetV2-SSD.
Mobilenet v2 is an improvement over mobilenet v1, which proposes two new concepts: inverted residual Inverted Residual and linear bottleneck Linear Bottleneck. The inverted residual Inverted Residual is mainly used to increase the extraction of image features to improve accuracy, while the linear bottleneck Linear Bottleneck is mainly used to avoid the information loss of the nonlinear function ReLU. The core of mobilenet v2 consists of 17 boltlenecks whose network structure is shown in table 3 (where t is the multiple of the up-scaling in the Bottleneck layer, c is the dimension of the output feature, n is the number of repetitions, s is the step size of the convolution, and k is the width scaling factor).
Table 3 MobileNetV2 network structure table
The specific structure of the bottleneck layer is shown in table 4. The input increases the dimension from k dimension to tk dimension through the 1 x 1 conv+relu layer, then downsamples the image by 3 x 3conv+relu separable convolution (when stride > 1), where the feature dimension is already tk dimension, and finally decreases the dimension from tk to k' dimension by 1 x 1conv (no ReLU).
Table 4 detailed structure table of bottleneck layer
Furthermore, for the bottleneck layer, when stride=1, the sum of elementwise is used to connect the input and output features (as in fig. 3); when stride=2, no shortcut connects the input and output features (as in fig. 4).
SSD is a single-stage object detection algorithm that utilizes feature maps of different scales to predict objects of different frame sizes. The SSD network structure is divided into two parts: basic network + pyramid network, wherein the basic network is transformable. The base network of the original SSD is the first 4-layer network of VGG-16, and the pyramid network is a simple convolution network with a feature map that tapers down, consisting of 5 parts. The specific network structure of the SSD is shown in FIG. 5.
The MobileNet V2-SSD deep learning neural network uses the MobileNet V2 network to replace VGG-16 in the original SSD network architecture, the configuration from Conv0 to Conv13 is completely consistent with the MobileNet V2 model, except that the final global average pooling, full connection layer and Softmax layer of the MobileNet V2 are removed, and the FC6 and FC7 of the original VGG-16 are replaced by Conv6 and Conv7 respectively. The MobileNet V2-SSD deep learning neural network firstly uses the MobileNet V2 network to extract image characteristic output characteristic graphs, and then uses an SSD target detection algorithm to detect information on a plurality of characteristic graphs output by the MobileNet V2 network.
In one embodiment, the ReLU_6 function is used as an activation function in training the target detection model.
To avoid overfitting, the network learning rate is set by adopting an exponential decay method, and an L2 regularization method is adopted (L2 regularization refers to square sums of elements in weight vectors and then square roots of the elements) based on an L2 norm, namely, adding an L2 norm sum term of parameters, namely, square sums of parameters and product terms of parameters, namely, the product terms of the square sums of parameters, to an objective function:
wherein C is 0 Representing the original cost function, n is the number of samples, λ is the regularization term coefficient, and the regularization term is weighted against C 0 The specific gravity of the term, w, is the weight. The term following the plus sign in the formula is the L2 regular term.
In the L2 regularization, the model parameters are updated using the following formula:
in this embodiment, in the target detection process, the image is supervised and trained, and each signboard image has a corresponding tag and a prediction frame, so that model parameters are trained according to the tag and the prediction frame, and the final recognition accuracy is judged according to the tag and the prediction frame. When the training network passes through a plurality of iterations, the predicted value continuously converges to the error direction of the label and the predicted frame, and then the back propagation updates the parameters into each layer according to the chain rule. And each iteration can reduce propagation errors as much as possible according to the gradient descent optimization direction, and finally, the final target detection result of all the signboard images in the data set is obtained. In the invention, a model with highest target detection accuracy on a test set is taken as a target detection model, in the actual operation of an industrial scene, firstly, a picture is acquired in real time through a camera, a single-area water pit signboard image is taken as an input, the model automatically processes the image to identify the characteristics of the signboard, the characteristics of the signboard are predicted, and finally, a target detection result is output. According to the condition of the target detection model identification signboard, whether the water accumulation pit at the bottom of the heating furnace is abnormal or not is judged. And if the final model identifies the water accumulation pit signboard, the explanation is normal. If the final model does not recognize the water accumulation pit signboard, the water accumulation pit signboard is abnormal, and the water overflow in the water accumulation pit at the bottom of the heating furnace is indicated, so that the timely alarm of the abnormal condition of the water accumulation pit at the bottom of the heating furnace is realized.
Since the embodiments of the apparatus portion and the embodiments of the method portion correspond to each other, the contents of the embodiments of the apparatus portion are referred to the description of the embodiments of the method portion, and are not repeated herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical function division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory ((RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, etc.
The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications and variations of the invention be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.
Claims (6)
1. The method for identifying the abnormal water accumulation pit at the bottom of the heating furnace based on deep learning is characterized by comprising the following steps of:
acquiring a water accumulation pit image;
detecting the water accumulation pit image by using a trained target detection model to obtain a detection result; the detection result comprises a signboard and a non-signboard;
judging whether the water accumulation pit is abnormal or not according to the detection result, if the detection result is that the water accumulation pit comprises a signboard, judging that the water accumulation pit is not abnormal, otherwise, judging that the water accumulation pit is abnormal;
carrying out feature extraction on data in a training sample set by using a neural network based on deep learning to obtain a feature map, and detecting the feature map to obtain a detection result;
in the process of extracting the feature map, the neural network based on deep learning carries out inversion residual error processing and linear bottleneck processing on the data in the training sample set;
the neural network based on deep learning is MobileNet V2-SSD;
in training the target detection model, the ReLU_6 function is used as an activation function, and L2 regularization is used.
2. The deep learning-based heating furnace bottom pit anomaly identification method of claim 1, wherein the method of training the target detection model comprises:
acquiring a training sample set;
training the neural network based on deep learning based on the training sample set to obtain the target detection model.
3. The deep learning-based heating furnace bottom pit anomaly identification method of claim 2, wherein the data in the training sample set is preprocessed, the preprocessing including at least one of: normalization processing and data enhancement;
the data enhancement includes at least one of: clipping, flipping, rotation, change in brightness, change in contrast, change in saturation.
4. The method for identifying the abnormal water pit at the bottom of the heating furnace based on the deep learning according to claim 2, wherein the neural network based on the deep learning is trained by adopting a supervised training method.
5. Heating furnace bottom water accumulation pit anomaly identification system based on degree of depth study, its characterized in that includes:
the image acquisition module is used for acquiring a water accumulation pit image;
the result detection module is used for detecting the water accumulation pit image by using a trained target detection model based on deep learning to obtain a detection result;
the abnormality judging module is used for judging whether the water accumulation pit is abnormal or not according to the detection result;
carrying out feature extraction on data in a training sample set by using a neural network based on deep learning to obtain a feature map, and detecting the feature map to obtain a detection result;
in the process of extracting the feature map, the neural network based on deep learning carries out inversion residual error processing and linear bottleneck processing on the data in the training sample set;
the neural network based on deep learning is MobileNet V2-SSD;
in training the target detection model, the ReLU_6 function is used as an activation function, and L2 regularization is used.
6. An electronic device, comprising:
a processor; and
a machine readable medium having instructions stored thereon, which when executed by the processor, cause the apparatus to perform the method of any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010101729.3A CN111310671B (en) | 2020-02-19 | 2020-02-19 | Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010101729.3A CN111310671B (en) | 2020-02-19 | 2020-02-19 | Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111310671A CN111310671A (en) | 2020-06-19 |
| CN111310671B true CN111310671B (en) | 2023-04-28 |
Family
ID=71147313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010101729.3A Active CN111310671B (en) | 2020-02-19 | 2020-02-19 | Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111310671B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037197A (en) * | 2020-08-31 | 2020-12-04 | 中冶赛迪重庆信息技术有限公司 | Hot-rolled bar cold-shearing material accumulation detection method, system and medium |
| CN112115877A (en) * | 2020-09-21 | 2020-12-22 | 四川弘和通讯有限公司 | Equipment abnormity identification method based on separable convolutional neural network |
| CN112580581A (en) * | 2020-12-28 | 2021-03-30 | 英特灵达信息技术(深圳)有限公司 | Target detection method and device and electronic equipment |
| CN112884036A (en) * | 2021-02-09 | 2021-06-01 | 北京京能能源技术研究有限责任公司 | Boiler heating surface abnormal image identification method, marking method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448382A (en) * | 2018-12-20 | 2019-03-08 | 天津天地伟业信息系统集成有限公司 | A kind of road depth of accumulated water monitoring and pre-alarming method |
| CN110472636A (en) * | 2019-07-26 | 2019-11-19 | 四创科技有限公司 | Water gauge E font scale recognition methods based on deep learning |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130184584A1 (en) * | 2012-01-17 | 2013-07-18 | Richard E. Berkey | Systems and methods for computerized ultrasound image interpretation and labeling |
| CN205879302U (en) * | 2016-07-18 | 2017-01-11 | 华南农业大学 | Water -level measuring post suitable for utilize image recognition water level |
| CN108344392B (en) * | 2017-01-23 | 2020-06-09 | 长城汽车股份有限公司 | Low-lying road surface information detection method and system and vehicle |
| US11187223B2 (en) * | 2017-04-10 | 2021-11-30 | Logical Concepts, Inc. | Home flood prevention appliance system |
| US11022124B2 (en) * | 2017-04-10 | 2021-06-01 | Logical Concepts, Inc. | Whole home water appliance system |
| CN108509919B (en) * | 2018-04-03 | 2022-04-29 | 哈尔滨哈船智控科技有限责任公司 | Method for detecting and identifying waterline in video or picture based on deep learning |
| CN108985322B (en) * | 2018-06-01 | 2019-11-22 | 广东电网有限责任公司 | A kind of cable tunnel ponding positioning identifying method based on ZF-Faster RCNNs |
| CN109543596B (en) * | 2018-11-20 | 2022-06-07 | 浙江大华技术股份有限公司 | Water level monitoring method and device, electronic equipment and storage medium |
| CN109670404B (en) * | 2018-11-23 | 2023-07-11 | 江苏理工学院 | Road ponding image detection early warning method based on hybrid model |
-
2020
- 2020-02-19 CN CN202010101729.3A patent/CN111310671B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448382A (en) * | 2018-12-20 | 2019-03-08 | 天津天地伟业信息系统集成有限公司 | A kind of road depth of accumulated water monitoring and pre-alarming method |
| CN110472636A (en) * | 2019-07-26 | 2019-11-19 | 四创科技有限公司 | Water gauge E font scale recognition methods based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111310671A (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111310671B (en) | Heating furnace bottom water accumulation pit anomaly identification method, system and equipment based on deep learning | |
| Ali et al. | Structural crack detection using deep convolutional neural networks | |
| CN108564085B (en) | Method for automatically reading of pointer type instrument | |
| CN113159147B (en) | Image recognition method and device based on neural network and electronic equipment | |
| Rahaman et al. | An efficient multilevel thresholding based satellite image segmentation approach using a new adaptive cuckoo search algorithm | |
| CN103593670A (en) | Copper sheet and strip surface defect detection method based on-line sequential extreme learning machine | |
| CN114758249B (en) | Target object monitoring method, device, equipment and medium based on field night environment | |
| CN112926652B (en) | Fish fine granularity image recognition method based on deep learning | |
| CN111339902A (en) | Liquid crystal display number identification method and device of digital display instrument | |
| Ngugi et al. | A new approach to learning and recognizing leaf diseases from individual lesions using convolutional neural networks | |
| CN113283446A (en) | Method and device for identifying target object in image, electronic equipment and storage medium | |
| CN113569852A (en) | Training method and device of semantic segmentation model, electronic equipment and storage medium | |
| CN111950812A (en) | Method and device for automatically identifying and predicting rainfall | |
| CN113449811A (en) | Low-illumination target detection method based on MS-WSDA | |
| CN110276759B (en) | Mobile phone screen bad line defect diagnosis method based on machine vision | |
| Dong et al. | Automatic inspection and evaluation system for pavement distress | |
| Lin et al. | An antagonistic training algorithm for TFT-LCD module mura defect detection | |
| CN108520261A (en) | A kind of recognition methods of peanut kernels quantity and device | |
| CN117274278B (en) | Retina image focus part segmentation method and system based on simulated receptive field | |
| CN111724338A (en) | Turntable abnormity identification method, system, electronic equipment and medium | |
| CN116129100A (en) | Truck part positioning and fault identifying method, device, equipment and medium | |
| CN113255440B (en) | Crop leaf abnormity detection method and system based on machine learning | |
| CN115439850A (en) | Image-text character recognition method, device, equipment and storage medium based on examination sheet | |
| CN115019306A (en) | Embedding box label batch identification method and system based on deep learning and machine vision | |
| CN114529906A (en) | Method and system for detecting abnormity of digital instrument of power transmission equipment based on character recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing Applicant after: MCC CCID information technology (Chongqing) Co.,Ltd. Address before: Building 1, No. 11, Huijin Road, North New District, Yubei District, Chongqing Applicant before: CISDI CHONGQING INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |